Provably Robust Detection of Out-of-distribution Data (almost) for free

• URL: http://arxiv.org/abs/2106.04260v1
• Date: Tue, 8 Jun 2021 11:40:49 GMT
• Title: Provably Robust Detection of Out-of-distribution Data (almost) for free
• Authors: Alexander Meinke, Julian Bitterwolf, Matthias Hein
• Abstract summary: Deep neural networks are known to produce highly overconfident predictions on out-of-distribution (OOD) data. In this paper we propose a novel method where from first principles we combine a certifiable OOD detector with a standard classifier into an OOD aware classifier. In this way we achieve the best of two worlds: certifiably adversarially robust OOD detection, even for OOD samples close to the in-distribution, without loss in prediction accuracy and close to state-of-the-art OOD detection performance for non-manipulated OOD data.
• Score: 124.14121487542613
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: When applying machine learning in safety-critical systems, a reliable assessment of the uncertainy of a classifier is required. However, deep neural networks are known to produce highly overconfident predictions on out-of-distribution (OOD) data and even if trained to be non-confident on OOD data one can still adversarially manipulate OOD data so that the classifer again assigns high confidence to the manipulated samples. In this paper we propose a novel method where from first principles we combine a certifiable OOD detector with a standard classifier into an OOD aware classifier. In this way we achieve the best of two worlds: certifiably adversarially robust OOD detection, even for OOD samples close to the in-distribution, without loss in prediction accuracy and close to state-of-the-art OOD detection performance for non-manipulated OOD data. Moreover, due to the particular construction our classifier provably avoids the asymptotic overconfidence problem of standard neural networks.

Related papers

• Mitigating Overconfidence in Out-of-Distribution Detection by Capturing Extreme Activations [1.8531577178922987]
  "Overconfidence" is an intrinsic property of certain neural network architectures, leading to poor OOD detection. We measure extreme activation values in the penultimate layer of neural networks and then leverage this proxy of overconfidence to improve on several OOD detection baselines. Compared to the baselines, our method often grants substantial improvements, with double-digit increases in OOD detection.
  arXiv  Detail & Related papers  (2024-05-21T10:14:50Z)
• Distilling the Unknown to Unveil Certainty [66.29929319664167]
  Out-of-distribution (OOD) detection is essential in identifying test samples that deviate from the in-distribution (ID) data upon which a standard network is trained. This paper introduces OOD knowledge distillation, a pioneering learning framework applicable whether or not training ID data is available.
  arXiv  Detail & Related papers  (2023-11-14T08:05:02Z)
• Can Pre-trained Networks Detect Familiar Out-of-Distribution Data? [37.36999826208225]
  We study the effect of PT-OOD on the OOD detection performance of pre-trained networks. We find that the low linear separability of PT-OOD in the feature space heavily degrades the PT-OOD detection performance. We propose a unique solution to large-scale pre-trained models: Leveraging powerful instance-by-instance discriminative representations of pre-trained models.
  arXiv  Detail & Related papers  (2023-10-02T02:01:00Z)
• Pseudo-OOD training for robust language models [78.15712542481859]
  OOD detection is a key component of a reliable machine-learning model for any industry-scale application. We propose POORE - POsthoc pseudo-Ood REgularization, that generates pseudo-OOD samples using in-distribution (IND) data. We extensively evaluate our framework on three real-world dialogue systems, achieving new state-of-the-art in OOD detection.
  arXiv  Detail & Related papers  (2022-10-17T14:32:02Z)
• Igeood: An Information Geometry Approach to Out-of-Distribution Detection [35.04325145919005]
  We introduce Igeood, an effective method for detecting out-of-distribution (OOD) samples. Igeood applies to any pre-trained neural network, works under various degrees of access to the machine learning model. We show that Igeood outperforms competing state-of-the-art methods on a variety of network architectures and datasets.
  arXiv  Detail & Related papers  (2022-03-15T11:26:35Z)
• Learn what you can't learn: Regularized Ensembles for Transductive Out-of-distribution Detection [76.39067237772286]
  We show that current out-of-distribution (OOD) detection algorithms for neural networks produce unsatisfactory results in a variety of OOD detection scenarios. This paper studies how such "hard" OOD scenarios can benefit from adjusting the detection method after observing a batch of the test data. We propose a novel method that uses an artificial labeling scheme for the test data and regularization to obtain ensembles of models that produce contradictory predictions only on the OOD samples in a test batch.
  arXiv  Detail & Related papers  (2020-12-10T16:55:13Z)
• Certifiably Adversarially Robust Detection of Out-of-Distribution Data [111.67388500330273]
  We aim for certifiable worst case guarantees for OOD detection by enforcing low confidence at the OOD point. We show that non-trivial bounds on the confidence for OOD data generalizing beyond the OOD dataset seen at training time are possible.
  arXiv  Detail & Related papers  (2020-07-16T17:16:47Z)
• Robust Out-of-distribution Detection for Neural Networks [51.19164318924997]
  We show that existing detection mechanisms can be extremely brittle when evaluating on in-distribution and OOD inputs. We propose an effective algorithm called ALOE, which performs robust training by exposing the model to both adversarially crafted inlier and outlier examples.
  arXiv  Detail & Related papers  (2020-03-21T17:46:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.